June j, 1905 J 



NA TURE 



117 



cxciv. pp. I to 36. The results here given differ from 

 those obtained by Mr. Grindley in one important particular. 

 Mr. Grindley came to the conclusion that steam taken 

 from a separator contained a definite proportion of 

 suspended moisture, because when he caused such steam 

 to e.Kpand through an orifice to a slightly reduced pressure 

 the steam did not become superheated, but its temperature 

 fell to that corresponding to saturated steam at the new 

 pressure. In the research here described, however, it was 

 found that steam tal^en from a separator and reduced in 

 pressure in the slightest degree by wire-drawing became 

 superheated. This result was only obtained after a con- 

 siderable amount of experimental work had been done, 

 and a number of improvements made in the apparatus as 

 first constructed. 



The experimental results obtained in the throttling ex- 

 periments are represented in the accompanying figure. 

 The curves a. b, c, d, e, and f show the connection 

 between the temperature and pressure of superheated 

 steam for constant total heat. The method of obtaining 



purposes of these calculations, and that a great degree of 

 accuracy would be necessary before such was the case. 



In method (2) the rise in temperature was observed in 

 steam flowing at a measured rate, due to the heat imparted 

 by an electric current, and the specific heat calculated 

 from the formula 



K„ — 



electrical input in watts XO'236 

 grams of steam passing per sec. X temp, rise ° C. 



10 ZO 50 40 50 60 70 80 90 100 110 120 



Absolute pressure 



each of these curves was as follows : — ^The pressure of 

 steam in the separator was maintained constant at the 

 point where the constant total heat curve meets the curve 

 which connects the pressure with temperature for saturated 

 steam ; the pressure on the low-pressure side of the orifice 

 could be regulated by means of a wheel valve, which 

 allowed the steam to escape at any desired rate. This 

 lower pressure was adjusted to various values, and the 

 temperatures corresponding were observed when the con- 

 ditions had become steady. By plotting these results 

 points were obtained which enabled the curve to be drawn. 

 The total heat corresponding to each of the curves was 

 obtained from Regnault's tables for saturated steam, and 

 the specific heat at constant pressure calculated for various 

 pressures. The specific heat as thus calculated was found 

 to increase rapidly with increase of temperature from 

 043 at 230° F. to 10 at 350° F. This apparent increase 

 in the specific heat led the author to suspect the accuracy 

 of Regnault's tables, and caused him to turn to the direct 

 heating method, with the result that he is now convinced 

 that Regnault's tables are not sufficientlv accurate for the 



NO. 1857, VOL. 72] 



The connection between grams of steam passing per 

 minute and the input of electrical energy in watts for a 

 definite rise in temperature was obtained for rates of 

 flow differing over a considerable range; the points thus 

 obtained were plotted on squared paper, and were found 

 to lie on a straight line which did not pass through the 

 origin, but cut the watts ordinate at a height correspond- 

 ing to the radiation loss expressed in watts. 



The difficulty experienced in keeping all the conditions 

 constant during the long time necessary for a complete 

 set of points was always 

 considerable. Numerous 

 experiments were carried 

 out, but the results varied 

 too much amongst them- 

 selves to enable con- 

 clusions to be drawn as to 

 the manner in which the 

 specific heat may vary 

 with pressure or tempera- 

 ture, except that any such 

 variation must be small, 

 and by no means of the 

 order indicated by the re- 

 sults of the throttling ex- 

 periments based on Reg- 

 nault's tables. 



The mean value of the 

 specific heat of super- 

 heated steam at constant 

 pressure obtained from the 

 most satisfactory experi- 

 ments was 0-46. 



May II. — "A Study of 

 the Process of Nitrifi- 

 cation with Reference to 

 the Purification of Sew- 

 age." By Dr. Harriette 

 Chick. Communicated by 

 Prof. Marshall 'Ward, 

 F.R.S. 



The process of nitrifi- 

 cation during sewage puri- 

 fication was studied by 

 means of small experi- 

 mental filters erected in 

 the institutes of hygiene in 

 \'ienna and Munich. 



The oxidation of sewage 

 passing through the filters 

 was investigated during the maturing period, and also when 

 the filters were mature, a special study being made, 

 chemically, of the oxidation of the nitrogen from the 

 ammoniacal form to that of nitrites and nitrates, and of 

 the distribution of these processes both in time and space. 



Nitrification was traced to the activity of two sets of 

 organisms, the first of which oxidised ammonia to nitrous 

 acid, and the second completed the oxidation to nitric acid. 

 These bacteria were found to differ only very slightly from 

 those isolated from the soil by Winogradsky, thus confirm- 

 ing the recent results of Schultz-Schultzenstein. The 

 activity in sewage filters of these organisms, which are 

 very sensitive to the presence of organic matter, requires 

 explanation, and various explanations are considered, based 

 upon experimental foundation. 



The theory of previous physical absorption of ammonia 

 upon the surface of the filtering material and subsequent 

 nitrification was found to be unsupported by experiment ; 

 nitrification is rather to be considered as a very rapid 

 biological process, requiring only the time taken by the 

 sewage to trickle through the filter. 



150 160 170 180 190 ZflO ZIO ZZO 



lbs. per square inch. 



